Hacker News: TonyAlicea10

New comment by TonyAlicea10 in "Issue: Claude Code is unusable for complex engineering tasks with Feb updates"

TonyAlicea10 — Mon, 06 Apr 2026 19:19:15 +0000

I've noticed this as well. "Now you should stop X and go do Y" is a phrase I see repeated a lot. Claude seems primed to instruct me to stop using it.

New comment by TonyAlicea10 in "Why AI-Assisted Development Is More Exhausting Than It Should Be"

TonyAlicea10 — Thu, 26 Mar 2026 13:54:04 +0000

1986. Shows how little has truly changed. I think this quote from that paper is one of the most apt for the LLM age: "The hardest single part of building a software system is deciding precisely what to build."

Why AI-Assisted Development Is More Exhausting Than It Should Be

TonyAlicea10 — Thu, 26 Mar 2026 13:36:21 +0000

Article URL: https://tonyalicea.dev/blog/single-mode-burnout/

Comments URL: https://news.ycombinator.com/item?id=47530302

Points: 3

# Comments: 3

New comment by TonyAlicea10 in "I built an AI receptionist for a mechanic shop"

TonyAlicea10 — Tue, 24 Mar 2026 11:29:38 +0000

“No hallucinations allowed” unfortunately suggests a misunderstanding of how LLMs work. It will work great until it doesn’t.

Trace: A Declarative Modeling Spec Format for the AI Age

TonyAlicea10 — Wed, 18 Mar 2026 13:32:09 +0000

Article URL: https://tonyalicea.dev/blog/trace-declarative-modeling-ai-age/

Comments URL: https://news.ycombinator.com/item?id=47425596

Points: 1

# Comments: 0

New comment by TonyAlicea10 in "Agents that run while I sleep"

TonyAlicea10 — Tue, 10 Mar 2026 21:17:53 +0000

You can find approaches that improve things, but there's always going to be a chance that your code is terrible if you let an LLM generate it and don't review it with human eyes.

But review fatigue and resulting apathy is real. Devs should instead be informed if incorrect code for whatever feature or process they are working on would be high-risk to the business. Lower-risk processes can be LLM-reviewed and merged. Higher risk must be human-reviewed.

If the business you're supporting can't tolerate much incorrectness (at least until discovered), than guess what - you aren't going to get much speed increase from LLMs. I've written about and given conference talks on this over the past year. Teams can improve this problem at the requirements level: https://tonyalicea.dev/blog/entropy-tolerance-ai/

The Future of Self-Paced Online Education

TonyAlicea10 — Tue, 24 Feb 2026 13:21:22 +0000

Article URL: https://tonyalicea.dev/blog/the-future-of-self-paced-online-education/

Comments URL: https://news.ycombinator.com/item?id=47136804

Points: 1

# Comments: 0

New comment by TonyAlicea10 in "JavaScript-heavy approaches are not compatible with long-term performance goals"

TonyAlicea10 — Mon, 16 Feb 2026 11:21:39 +0000

React compiler adds useMemo everywhere, even to your returned templates. It makes useMemo the most common hook in your codebase, and thus very necessary. Just not as necessary to write manually.

New comment by TonyAlicea10 in "Ask HN: Share your personal website"

TonyAlicea10 — Wed, 14 Jan 2026 20:48:02 +0000

https://tonyalicea.dev

New comment by TonyAlicea10 in "Coursera to combine with Udemy"

TonyAlicea10 — Wed, 17 Dec 2025 20:14:56 +0000

I've been an instructor on Udemy for a decade. Lots of web devs tell me they got their career started on my courses there, which is something that is very dear to me. Udemy got me started in online dev education.

I'm very sorry to see it go.

New comment by TonyAlicea10 in "PGlite – Embeddable Postgres"

TonyAlicea10 — Thu, 04 Dec 2025 12:42:41 +0000

One thing this (and any browser-embeddable data store) are fantastic for is vibe coding interactive prototypes for user research.

AI-generated code that doesn’t need to be production ready has been a real boon to usability and design work. Testing with users something that actually saves and displays data, and seeding the app with realistic-looking datasets in both shape and size, reveals usability issues that you just don’t discover in Figma prototypes.

If a product team isn’t performing user research with interactive prototypes as a core part of their dev and design lifecycle, they’re doing themselves a real disservice. It’s so easy now.

New comment by TonyAlicea10 in "Confessions of a Software Developer: No More Self-Censorship"

TonyAlicea10 — Sat, 29 Nov 2025 12:55:28 +0000

This article is brave. It is also is an example of a general issue: we assume our personal experiences are representative when they are, in fact, unique.

“Follow Scrum, Lean / Kanban, or eXtreme Programming to the letter” - there are plenty of failed projects and unhappy devs that have done just that. And these methodologies are not tuned for the LLM-generation age and, talking to lots of other devs around the world, I think it is showing.

In regards to remote work, I’ve worked for shops that have been fully remote since before the pandemic and are wonderful experiences. They’ve figured it out. The OP’s feelings on remote work, to me, say “the companies I’ve worked for are really bad at supporting remote work”, but if you believe your experiences are representative then you say “remote work is bad”.

New comment by TonyAlicea10 in "We're losing our voice to LLMs"

TonyAlicea10 — Thu, 27 Nov 2025 18:58:44 +0000

I'm the OP. I can attest that I am not an LLM model creator! :)

I consider myself an LLM pragmatist. I use them where they are useful, and I educate people on them and try to push back on all the hype marketing disguised as futurism from LLM creators.

New comment by TonyAlicea10 in "We're losing our voice to LLMs"

TonyAlicea10 — Thu, 27 Nov 2025 15:14:39 +0000

I think there's a difference between using an LLM as an editor and asking the LLM to write something for you. The output in the former I find to still have a far clearer tonal fingerprint than the latter.

And whose to say your idiosyncratic expressions wouldn't find an audience as it changes over time? Just you saying that makes me curious to read something you wrote.

We're losing our voice to LLMs

TonyAlicea10 — Thu, 27 Nov 2025 14:51:01 +0000

Article URL: https://tonyalicea.dev/blog/were-losing-our-voice-to-llms/

Comments URL: https://news.ycombinator.com/item?id=46069771

Points: 354

# Comments: 402

Stop Adding AI to Bad Process [video]

TonyAlicea10 — Wed, 26 Nov 2025 13:45:29 +0000

Article URL: https://youtu.be/gCdM4hdwBmY?si=8M6khbUEIeX4XB_5

Comments URL: https://news.ycombinator.com/item?id=46057335

Points: 1

# Comments: 0

New comment by TonyAlicea10 in "[dead]"

TonyAlicea10 — Thu, 16 Jan 2025 22:13:58 +0000

Tools like bolt.new from StackBlitz and Cursor have brought about a new way to prototype that has some distinct advantages over traditional tools like Figma.

This blog post gives 9 principles to guide what I'm calling "Generative Rapid Prototyping" (or GRiP).

New comment by TonyAlicea10 in "[dead]"

TonyAlicea10 — Tue, 14 Jan 2025 19:19:48 +0000

A deep dive into the internals of React and NextJS to form an accurate mental model of RSCs.

How to Rewrite a Web Application

TonyAlicea10 — Mon, 29 Jul 2024 20:19:40 +0000

Article URL: https://tonyalicea.dev/blog/how-to-rewrite-a-web-application/

Comments URL: https://news.ycombinator.com/item?id=41103388

Points: 5

# Comments: 0

New comment by TonyAlicea10 in "Smart jerks aren't worth it"

TonyAlicea10 — Thu, 25 Jul 2024 03:12:34 +0000

No, not necessarily. I’m referring to consistent behaviors across a wide range of circumstances and contexts.